Pro & Con

CHARLES K. FRANCIS, M.D., is the iImmediate past president of the American College of Physicians.

BRUCE VLADECK is a principal at Ernst & Young L.L.P., and the former administrator of the Health Care Financing Administration.

Yes | No

Yes - by Charles K Franis, M.D. Health care in the United States is currently fragmented and uncoordinated across different practice settings. Performance measurement systems present an opportunity to establish evidence-based norms of care. In addition, valid and accepted indicators can be used as proxies to reflect the general state of health care. By pointing to disparities in care, indicators have the potential to raise the overall floor of acceptable practice standards.

Evaluating physician performance to improve quality of care is not a new subject. What has changed is the government's and market's demand for better information on which to make health care choices.

The impact has been a heightened demand for comparative provider performance data, helping to raise the bar on quality by bringing professional accountability to the health care marketplace.

The ultimate goal of performance measures is to create a system that ensures the highest quality of care through transparency, accountability, and credibility. The goal is not to punish, but to incentivize improvements in care. To ensure that any performance measurement systems established meet this goal, the American College of Physicians feels the following positions must be kept in mind:

PIThe goal of physician performance measurement should be to foster continuous quality improvement of clinical care to meet or exceed evidence-based national standards of such care.

PIPhysician performance measures should be evidence based, broadly accepted, and clinically relevant. These measures should assess and focus on those elements of clinical care over which physicians have direct control (as opposed to systems constraints). They should be built on statistical methods that provide valid and reliable comparative assessment across populations.

PIAny data collection required to support performance measurement should be feasible, reliable, affordable, and practical. Data collection should not violate patient privacy or add to the paperwork burden experienced by physicians. Information technology tools are crucial to this data collection, and incentives need to be added to make sure physicians are able acquire these tools.

PIThe ACP supports demonstration projects to evaluate the use of incentives, including financial incentives, to reward physicians who meet or exceed performance standards. Any financial incentives related to performance measurement should be directed at positive rather than negative reward.

With respect to the last position, the ACP has urged Congress to expand the small practice pay-for-performance demonstration project created by the Medicare Modernization Act of 2003. This project is designed to aid practices in the acquisition of health information technology.

The project needs to be expanded to include a much larger number of states beyond the four states authorized by Section 649 and to a larger number of physician practices in each state. It also needs to allow for all practices that participate in the program to be eligible for incentives.

If done correctly, performance measures have a great potential to assess physician performance, to improve the quality of patient care, to enhance the coordination and management of care, and to reward physicians who meet or exceed the benchmarks.

NO - by BRUCE VLADECK.  People today are appropriately concerned about improving the quality of health care in the United States, but I think pay for performance is going to make it harder for us to be able to get there.

While I do believe in administered pricing in public programs, I don't believe you can achieve every social objective by "pricing" quality of care.

Many areas of health care would stand to improve if people had better infrastructure and information technology resources, such as electronic medical records.

But from the point of view of an individual physician's office or from that of a small provider, it's not economically worthwhile to make that investment.

And we're never going to get that investment unless there's some public subsidy to do it.

To suggest, however, that the way to generate that subsidy is to add $5 to every office visit in the Medicare program, makes no sense.

Any rational provider would pocket the $5 and then not buy the information technology.

People who have focused on efforts to measure and improve quality of care for long periods of time are often frustrated by the fact that we haven't "solved the problem" yet.

Washington in particular tends to believe in the magical power of financial incentives to change any form of human behavior. Get the incentives right; get the behavior right.

It's an issue I've been engaged in for the better part of the last 25 years. In 1979 I became a principal investigator on a major Health Care Financing Administration contract to implement a diagnosis-related groups (DRG)-based hospital payment system in New Jersey. The initial objective of the DRG program was to find a way to pay for quality.

We were going to develop a unit of payment for inpatient hospital services that could also be used as a way of measuring comparative quality, and for building the appropriate incentives and the disincentives needed for the quality of care into the payment system.

As the project proceeded over the years, we discovered that DRGs were a pretty good tool for paying the hospitals--but they turned out to be a lousy tool whith which to measur quality.

We also found, in development of all kinds of hospital payment policies, that there were enormous variations from hospital to hospital in the quality of services.

But we could never answer the critical question: If hospitals had low costs but low quality, should we pay them more or pay them less?

While it's true that physicians and other clinicians and the people around hospitals are under economic pressure, that's not what they're solely about. Most want to do a good job.

To change behavior through relatively modest economic incentives is in my opinion much less powerful than targeting the activities that make professionals feel good about themselves--where their identity is closely tied.

Hiring a hospitalist to get 1% extra on a Medicare payment for DRGs is the wrong calculus to ask people to make. Speaking to the values and professional expectations of the health professionals throughout the system is a better way to improve quality of care.